Mapping the locations where transcription factors interact with DNA is central to uncovering genome regulation mechanisms and gives key insights as to how organisms respond to intracellular or extracellular signals. The current gold standard for characterizing transcription factor binding sites is Chromatin Immunoprecipitation Sequencing (ChIP-Seq). Despite the ubiquity of ChIP-Seq, significant caveats remain. This method requires high-quality antibodies that may be laborious and costly to acquire, and large amounts of material are needed to obtain reliable data. We propose an alternative to ChIP-Seq: exploiting the interbacterial toxin double-stranded DNA deaminase (DddA). DddA catalyzes the deamination of cytosine to uracil, which is replaced by thymine via DNA replication. We harness this activity by designing translational fusions of DddA to the protein of interest. We hypothesize that the target protein carries DddA to its DNA binding region, introducing local C→T mutations that deep-sequencing detects. As a proof of concept, we tested this system with known bacterial transcription factors GacA, GcsR, and FleQ. Our results confirmed that DddA fusions have higher localized C→T mutations where the characterized transcription factors bind. We also tested different conditions and construct designs to maximize the efficiency of this method. Our results show that the use of DddA as a method of analyzing protein–DNA interactions is a broad and powerful tool with advantages over current methods like ChIp-Seq. Through this, we can better understand how organisms control gene expression when infecting a host, competing for resources, or surviving stressors.

Many human diseases, including cancer metastasis, and developmental disorders, involve sheets of cells moving together. This concerted process, called collective cell migration (CCM), is critical to maintaining the integrity of the sheet of cells during movement. Maintaining cell sheet integrity is dependent on individual cells retaining their shape while still connected to neighboring cells. To study collective cell migration, we use Drosophila melanogaster (fruit fly) embryogenesis. Dorsal closure (DC), an event in Drosophila embryogenesis, is an example of CCM. During DC, two epithelial layers wrap around the embryo. Because CCM’s core mechanics are conserved, studying DC provides an opportunity to indirectly study human diseases/disorders function in a model that has fewer genetic redundancies and is easy to manipulate. Adherens junctions (AJs), are the site of adhesion between individual cells, and what allows for dynamic rearrangements like CCM by strategically increasing or decreasing adhesion. Despite the importance of this process and its relevance to many diseases, CCM is highly complex and how the scaffolding protein Canoe stabilizes adherens junctions under high stress is not entirely understood. We are interested in how junctions between cells are stabilized during a dynamic movement like in CCM. Using a maternally driven Gal4/UAS system, we created moderate canoe knockdowns. We then performed targeted, immunofluorescence antibody staining of embryos during DC. We expect that the loss of Canoe will lead to increases in cell-sheet tension as well as the compensation by other stabilizing proteins like Polychaetoid. Our research improves our understanding of stabilizing proteins at AJs and CCM and may lead to the more effective treatment of disrupted CCM human diseases/disorders like wound healing, palate formation, and neural tube closure.

More than one in 5,000 individuals are affected by mitochondrial dysfunction, which is considered a hallmark of aging by being associated with cancer, diabetes, and Alzheimer’s disease. A better understanding of the pathophysiology of mitochondrial disease progression could potentially lead to the discovery of novel interventions to treat these diseases. The Kaeberlein Lab uses a mouse strain that is deficient in a Complex I subunit of the Electron Transport Chain (NDUFS4) as a model of mitochondrial disease. These mice exhibit symptoms including retarded growth, loss of motor activity, and neuroinflammation, eventually leading to premature death. As Complex I is normally composed of multiple iron-sulfur clusters, we hypothesized its deficiency would lead to intracellular iron defects that drive disease progression. Excess body iron leads to oxidative stress and is associated with many age-related and neurodegenerative diseases. Our group conducted lifespan experiments with iron-control and iron-deficient diets. I observed a 12% increase in lifespan and a delay in the onset of neurodegenerative phenotypes in the low-iron cohorts. Total iron levels were quantified in the liver, brain, and other tissues by ICP-MS, in which I observed 3-fold increases in total liver iron, consistent with iron overload. Furthermore, I performed ferrozine assays to determine non-heme iron levels which similarly showed increased non-heme iron levels in livers of KO mice relative to WT mice. The liver produces hepcidin in response to iron-overload, which is the master regulator of iron absorption and utilization in the body. I quantified hepcidin levels by ELISA to better understand the regulation of iron absorption in KO mice based on liver iron status. These results may have broader implications in Leigh syndrome, but also potentially in other disorders resulting from mitochondrial dysfunction.

Herpes Simplex Virus 2 (HSV-2) causes lifelong recurrent infection with intermittent lesion formation, typically around the genitals or anus. Long-lasting changes to the quantity and composition of immune cell populations occur at the site of viral release and lesion presentation. This research applied chip-based cytometric staining and imaging to characterize local immune responses in human skin biopsy tissue from two study participants with differing rates of viral shedding. An additional tissue from one participant was used to investigate changes in immune cell populations after lesion healing. In total, 12 biomarkers were stained on each tissue section and composite images were created allowing for immunophenotypic analysis of cells in these samples. The 12-plex immunostaining of a single tissue section was made possible by sequential application of fluorescently labeled antibody stains separated by a photobleaching step to remove previous marker signals. As a result, complex immune cell phenotypes and spatial distributions could be observed in situ. Both lesion samples show full epidermal ulcer site penetration by neutrophils (CD15+). Aside from some CD8+ cell penetration observed in the infrequent shedder tissue, CD4+ and CD8+ cells are generally excluded from the ulcer site but appear in large numbers throughout the surrounding dermis. CD11c+ and CD123+ dendritic cell populations can readily be seen nearby CD4+ and CD8+ cells, likely presenting antigens from within the ulcer site. There is a marked absence of CD57+CD8+ T cells throughout the sample from the participant with frequent viral shedding. Healed tissue from a former lesion site showed a dramatic decrease in total Immune cells. Understanding the role of localized immune cells in combating viral shedding may influence treatments aimed at increasing an infected persons ability to rapidly eliminate escaping virus, preventing lesion formation, resulting in increased quality of life and decreased risk of transmission.

Mitochondria are multi-functional organelles that regulate calcium signaling, an important signaling pathway that controls cellular processes ranging from transcriptional regulation to vesicular secretion. Calcium ions enter the mitochondria through the Mitochondrial Calcium Uniporter (MCU) complex, a highly selective calcium uniporter controlling calcium flux across the inner mitochondrial membrane. Previous data suggests uniporter inhibition alters lipid metabolism and induces a qualitative increase in cytosolic lipid droplets. Stemming from this initial observation, we created two research goals for this project: 1) to quantify and characterize lipids in response to MCU inhibition and 2) aims to explore the role of known or novel lipid synthesis pathways in MCU-dependent lipid accumulation. To understand the role of MCU in cellular physiology, we used CRISPR technology to knockout the MCU gene (MCU KO) in HeLa cell lines. We quantified differential lipid accumulation in single cell images through confocal fluorescence microscopy. In addition, potential mechanisms were investigated by Western Blot analysis and quantitative Polymerase Chain Reaction (qPCR). Here we verify that inhibition of mitochondrial calcium flux is responsible for lipid accumulation, a phenotype rescued by exogenous MCU expression. This accumulation is attributed to the upregulation of a transcription factor – Nuclear Factor of Activated T cells 4 (NFATc4). Gene knockdown of NFATc4 rescues the MCU phenotype, strongly suggesting NFATc4 as a downstream signaling factor of MCU inhibition. In follow-up experiments, we seek to identify MCU-regulated gene expression and understand their roles in regulation of metabolism. Although the function of increased cellular lipid content remains elusive, we theorize it may play a role in mitochondria-to-cytosol stress communication. Futher characterization of lipid-mediated stress pathways may identify novel targets for conveying increased cell stress tolerance.

Performing computational molecular dynamics (MD) simulations of small-molecule systems has become one of the most prominently used methods in studies of molecular structure and behavior. However, MD is a computationally expensive and time-consuming methodology because of the requirement of computing detailed interactions among atom-atom pairs. There is great interest, therefore, in reducing the time and computational power needed to approximate real-world systems. Most commonly, such efforts have employed machine learning techniques to predict extensive properties of molecular systems. Here, we propose accelerating simulations by predicting conformational changes - a prospect that has not yet been fully explored. Previous work attempted applying a linear dynamical analysis algorithm named Dynamic Mode Decomposition to MD data, which has been shown to be ineffective through a multiresolution analysis. We propose herein the use of Sparse Identification of Nonlinear Dynamical Systems (SINDy), a nonlinear model which has been shown to accurately decipher the governing equations of dynamical systems. We will be testing the effectiveness of SINDy with MD data by performing an iterative error analysis while varying the initial parameters of the dataset, thereby gaining a better understanding of how much data (and in what form) should be inputted to maximize the accuracy of a simulated SINDy model of an MD dataset. If shown to be sufficiently accurate, we then can implement SINDy simultaneously with MD in an active learning loop to save time and computational power while maintaining a high degree of predictive capability for peptide conformations. The current goal is to obtain a deeper understanding of peptide conformational changes that could, in the future, be combined with machine learning techniques to greatly accelerate classical MD simulations.

This project is supported by the UW Computational Neuroscience Center, and the DMREF Program of NSF through the MGI platform under DMR# 1629071, 1848911, and 1922020.